Broad, long-term objectives: The overexpression of epidermal growth factor receptor (EGFR) is observed in several types of solid tumors including non-small cell lung cancer and many carcinomas. We are using computational techniques to simulate binding of ligands known to interact with the ATP binding site on the kinase domain of EGFR in an effort to develop improved small molecules. Specifically, we are evaluating cancer causing EGFR mutations that include either a point mutation or a deletion on the activation loop that is positioned near the ATP binding site and the effect of these mutations on ligand binding. Other mutations, which arise as resistant mutations during treatment with compounds such as erlotinib and gefitinib will also be characterized. In order to develop small molecule inhibitors with specificity and high affinity for mutants, we will complete the following specific aims: (1) Characterize how mutations affect EGFR ligand binding using molecular dynamics followed by binding energy analysis. We will determine the behavior of mutations on ligand binding affinities, which amino acids in the binding pocket play an important role in the interaction between EGFR and each ligand, and the effect of each mutation on the active and inactive conformations of EGFR. (2) Design improved small molecule inhibitors based on computational analyses performed under aim #1 by ligand modification to improve affinity towards a specific mutation. We will use the scaffolds of the known inhibitors as well as that of the natural substrate to optimize binding affinity by performing functional r-group library searches. (3) Identify new lead structures using computational tools including virtual high throughput screening (docking) of ligand libraries. Again, we will optimize top scoring and populated structures using r-group library searches. The proposed aims will use the following computational methods: MD simulations, rigid and flexible docking using the anchor and grow algorithm;structural activity relationships and r-group libraries searches (rotamers). Methods for achieving the stated goals: The proposed aims use the following tools and software: AMBER, VMD, NAMD, MOE, Dock, BOSS, and BOMB. Relevance to Public health: Structure-based drug design is an important component of targeted drug development. The research proposed will characterize EGFR kinase domain mutations and their binding with known ligands using computer simulations. The use of the information generated will provide insight into the design of new ligands based on existing scaffolds and novel leads effective against EGFR mutants. Targeted design of new chemotherapeutic agents with high specificity and affinity will enable improved outcomes for patients with solid tumors that express mutant EGFR.